On the Python modeling API level, URDMEDataFunctions should behave like Parameters, i.e. it should be possible to do something like
self.add_data_function(MeshSize(self.mesh))
And then say:
p = pyurdme.Parameter(name="width",expression="L/h")
(where L is some other parameter and h is the str name of the datafunction). The code generated in the C level would in principle work as is now, but p would get the value None in the C file since it is not evaluable only in the namespace of the parameter list.
Solution: evaluate Parameters in a namespace that is a union of data functions and parameters prior to serializing to file.
On the Python modeling API level, URDMEDataFunctions should behave like Parameters, i.e. it should be possible to do something like
self.add_data_function(MeshSize(self.mesh))
And then say:
p = pyurdme.Parameter(name="width",expression="L/h")
(where L is some other parameter and h is the str name of the datafunction). The code generated in the C level would in principle work as is now, but p would get the value None in the C file since it is not evaluable only in the namespace of the parameter list.
Solution: evaluate Parameters in a namespace that is a union of data functions and parameters prior to serializing to file.